A control method of an optoelectronic tracking system based on an adaptive error observer

Abstract

The application discloses a photoelectric tracking system control method based on an adaptive error observer, mainly used for improving the tracking ability of a system to unknown change input signals and the suppression ability of the system to unknown time-varying narrow-band disturbances, and improving the tracking precision of the photoelectric tracking system. The application establishes an error observer structure to observe the sum signal of the input and the disturbance, then identifies the observation signal from the frequency domain, and finally adjusts the controller parameters in real time and on line, so as to reduce the tracking error of the system. The application can solve the tracking problem and the anti-disturbance problem of the system simultaneously, and under the condition that the input and the disturbance are unknown and change, the observation signal is identified and analyzed from the frequency domain, and then the controller parameters are adaptively changed, so that the tracking ability of the system to the moving target and the suppression ability of the system to the complex environmental disturbance are improved.

Description

Technical Field

[0001] This invention belongs to the field of tracking control, specifically relating to a control method for an optoelectronic tracking system based on an adaptive error observer. It is mainly used to cope with unknown changing inputs and unknown time-varying narrowband disturbances, thereby improving the tracking accuracy of the optoelectronic tracking system. Background Technology

[0002] In photoelectric tracking systems, tracking accuracy is easily affected by target motion characteristics and external disturbances, leading to a decrease in accuracy. On the one hand, the irregular motion of the target object and the difficulty in acquiring its trajectory limit the improvement of the system's tracking capability. On the other hand, photoelectric tracking equipment mounted on moving platforms such as aircraft and ships is also subject to low-frequency broadband and narrow-band disturbances transmitted from the carrier, thus affecting the system's tracking accuracy. In previous studies, tracking and anti-interference issues have typically been discussed separately. The paper "A Modified Disturbance Observer Structure Based on Acceleration Measurement for Disturbance Suppression in Tracking Control System" utilizes a disturbance observer structure to compensate for external disturbances in photoelectric tracking systems, effectively improving the system's anti-interference capability, but tracking performance remains unchanged. The paper "Error-Based Observer of a Charge Coupled Device Tracking Loop for Fast Steering Mirror" proposes an error observer structure that uses error information to observe the system's input and disturbances, simultaneously addressing tracking and anti-interference issues. However, this method uses fixed control parameters and cannot effectively handle unknown varying inputs and unknown time-varying narrowband disturbances, limiting tracking accuracy. To cope with varying inputs and disturbances, a real-time, online adaptive error observer method is needed. Summary of the Invention

[0003] To improve the tracking capability of photoelectric tracking systems to unknown changing input signals and the suppression capability to unknown time-varying narrowband disturbances, this invention proposes a control method for photoelectric tracking systems based on an adaptive error observer. In the position closed loop, an error observer structure is first established. Then, the fast Fourier transform method is used to identify the input and disturbance signals, and the main frequency components of the observed signals are acquired in real time. Next, based on the identification and analysis results, the parameters of a Q-filter composed of a low-pass filter and multiple notch filters connected in series are adjusted. Finally, the filtered observed signal is fed forward into the control loop, realizing the entire process of the adaptive disturbance observer. The Q-filter parameter adjustment is constrained by the stability of the small gain theorem. This invention overcomes the limitations of traditional methods, effectively improving the photoelectric tracking system's ability to respond to changing inputs and disturbances, thereby achieving higher tracking accuracy.

[0004] To achieve the objectives of this invention, a control method for a photoelectric tracking system based on an adaptive error observer is provided, the steps of which are as follows:

[0005] Step (1): Using the CCD as the position sensor of the photoelectric tracking system, the relative angular position between the target's line of sight and the sensor center is calculated, thus forming a position closed loop;

[0006] Step (2): In the position loop, the object characteristics of the position loop are obtained using a frequency response tester, and represented as object model G. p (s), where the input signal of the frequency response tester is the drive voltage input, and the output signal is the sampled output value of the CCD;

[0007] Step (3): Based on the object model G of the position loop p Design a position controller C(s) to complete closed-loop control;

[0008] Step (4): After completing the location closure, establish the object model G. p The nominal inverse model of (s) The error signal E(s) is input into the nominal inverse model. right The output signal is summed with the motor drive signal U(s) to obtain the sum estimate H(s) of the input signal R(s) and the disturbance signal D(s) in the position loop;

[0009] Step (5): Identify and analyze the main frequency components in H(s) using the Fast Fourier Transform method;

[0010] Step (6): Construct a Q filter, designing it as a low-pass filter and multiple notch filters connected in series, with variable filtering parameters;

[0011] Step (7): Based on the frequency domain identification and analysis results of the estimated quantity H(s), adjust the Q filter parameters to filter H(s), and then filter the signal. The output U fed forward to the controller C(s) C (s);

[0012] Step (8): Repeat steps (4) to (7) to complete the entire process of the adaptive error observer until the system finishes running.

[0013] Furthermore, in step (1), a CCD is used as a closed-loop sensor for the position loop to obtain the relative angular position between the target line of sight and the sensor center, thereby realizing a position feedback closed loop.

[0014] Furthermore, in step (2), the object characteristics of the position loop are approximated as a second-order oscillating element, then the object model G p (s) is expressed as follows:

[0015]

[0016] Where a, b, and K are model parameters.

[0017] Furthermore, in step (3), the position controller C(s) is designed using the zero-pole cancellation method.

[0018] Furthermore, in step (4), in order to obtain and estimate the quantity H(s), an error observer is established, and the error signal E(s) is input to the nominal inverse model. And the estimate H(s) is derived from The output signal is obtained by summing the motor drive signal U(s).

[0019] Furthermore, in step (5), the power spectrum of H(s) is obtained by using the fast Fourier transform method, and then the main frequency components in H(s) are identified and analyzed.

[0020] Furthermore, in step (6), the Q filter is composed of a low-pass filter and multiple notch filters connected in series, and its form is as follows:

[0021]

[0022] Where, β i , ε i T represents the filter parameters, f i Let be the dominant frequency of the i-th notch filter. To satisfy the stability condition, β must be... i ε i <1, and β i >1.

[0023] Furthermore, in step (7), the stability of the Q-filter is constrained by the small gain theorem:

[0024]

[0025] Furthermore, step (8) includes the following steps: first, constructing an error observer structure to obtain the sum estimate H(s) of the input signal R(s) and the disturbance signal D(s) in the loop; then, obtaining the main frequency components of H(s) through the fast Fourier transform method; subsequently, adaptively adjusting the notch filter parameters of the Q filter; and finally, filtering the signal... Feeding forward into the loop enhances the photoelectric tracking system's ability to track unknown changing input signals and its ability to suppress unknown time-varying narrowband disturbances.

[0026] Compared with the prior art, the present invention has the following advantages:

[0027] (1) Compared with position closed-loop control, the present invention adopts an observer structure without adding additional sensors, and can use algorithms to improve the tracking performance and disturbance suppression performance of the photoelectric tracking system.

[0028] (2) Compared with traditional control methods, this invention solves the separate tracking problem and anti-interference problem at the same time, providing a new idea for the research of control strategies for photoelectric tracking systems.

[0029] (3) Compared with the traditional error observer method, the present invention can adjust the control parameters in real time and online, thereby improving the system's ability to cope with unknown input changes and unknown time-varying disturbances.

[0030] (4) Compared with the traditional Q filter design method, the present invention adopts a low-pass filter and multiple notch filters in series, which can achieve better filtering performance while ensuring stability. Attached Figure Description

[0031] Figure 1 This is a control block diagram of a photoelectric tracking system control method based on an adaptive error observer according to the present invention.

[0032] Figure 2 This is a simulation result of the adaptive error observer method for photoelectric tracking systems. Detailed Implementation

[0033] The following diagram and accompanying drawings provide a detailed description of specific embodiments of the present invention.

[0034] As attached Figure 1The diagram shows a control block diagram of a photoelectric tracking system control method based on an adaptive error observer. It includes a CCD position closed loop, an error observer, and a sampling, identification, and analysis section for the estimated quantities. This method employs the error observer method, the fast Fourier transform method, and a novel Q-filter design approach. It can autonomously and in real-time estimate the sum of the input and disturbance signals. Then, through sampling and frequency domain identification analysis of the estimated quantities, it adaptively adjusts the Q-filter parameters, providing targeted feedforward for key frequency components. This improves the system's ability to track unknown changing inputs and suppress unknown time-varying disturbances. The specific implementation steps of the method are as follows:

[0035] Step (1) uses the CCD as the position sensor of the photoelectric tracking system, which can calculate the relative angular position between the target's line of sight and the sensor center, thus forming a position closed loop.

[0036] Step (2) In the position loop, the object characteristics of the position loop are obtained using a frequency response tester, and are represented as object mode G. p (s) The input signal of the frequency response tester is the driving voltage input, and the output signal is the sampled output value of the CCD. The object characteristics of the position loop can be approximated as a second-order oscillating element, and its model is expressed as follows:

[0037]

[0038] Where a, b, and K are model parameters.

[0039] Step (3) Based on the object model G of the position loop p (s), design position controller C(s) to complete closed-loop control.

[0040] After completing the position loop in step (4), establish the position loop object model G. p The nominal inverse model of (s) The error signal E(s) is input into the nominal inverse model. right By summing the output signal with the motor drive signal U(s), we can obtain the sum estimate H(s) of the input signal R(s) and the disturbance signal D(s) in the position loop.

[0041] Step (5) Use the Fast Fourier Transform (FFT) method to identify and analyze the main frequency components in H(s).

[0042] Step (6) Construct the Q-filter, designing it as a low-pass filter and multiple notch filters connected in series, with variable filtering parameters. The Q-filter consists of a low-pass filter and multiple notch filters connected in series, and its form is as follows:

[0043]

[0044] Where, β i , ε i T represents the filter parameters, f i Let be the dominant frequency of the i-th notch filter. To satisfy the stability condition, β must be... i ε i <1, and β i >1.

[0045] Step (7) Based on the frequency domain identification and analysis results of the estimated quantity H(s), adjust the Q filter parameters to filter H(s), and then filter the signal. The output U fed forward to the controller C(s) C (s), due to stability constraints, the filter parameters cannot change arbitrarily. The stability of the Q filter is constrained by the small gain theorem:

[0046]

[0047] Step (8) Repeat steps (4) to (7) to complete the entire process of the adaptive error observer.

[0048] In the position loop constructed by the CCD, an error observer structure is built. Using the Fast Fourier Transform (FFT) method, the main frequency components of the quantity H(s) are identified, analyzed, and estimated online. Based on the analysis results, the notch frequency of the Q-filter is adjusted in real time. Finally, the filtered signal is... Feeding forward into the loop enhances the photoelectric tracking system's ability to track unknown changing input signals and suppress unknown time-varying narrowband disturbances, thereby reducing tracking errors.

[0049] The following uses a practical photoelectric tracking system as an example to illustrate the design process and effects of this invention in detail:

[0050] (1) Use the attachment Figure 1 The control block diagram shown constructs an adaptive error observer structure, with the CCD serving as a position sensor.

[0051] (2) The frequency response tester measures the mathematical model G of the controlled object in the system position loop. p (s):

[0052]

[0053] (3) Based on object model G p A position controller C(s) is designed using the zero-pole cancellation method to achieve the tracking function.

[0054]

[0055] (4) To ensure It is physically feasible by adding two inertial elements. and After adding and The values ​​are almost the same within 50Hz and have no effect on the system.

[0056]

[0057] (5) In the error observer structure, the sum of the input and disturbance signals is estimated, and then the sum estimate is analyzed in the frequency domain by fast Fourier transform to obtain the main frequency components.

[0058] (6) Based on the frequency domain identification and analysis results, the notch parameters of the Q filter are adaptively adjusted to perform targeted filtering and restoration of multiple main frequency components, and then fed forward to the position loop to realize the entire process of the adaptive error observer. The Q filter structure is as follows:

[0059]

[0060] Among them, f i Let be the dominant frequency of the i-th notch filter, and the other parameters take the following values:

[0061] α i =5,η i =0.01, T = 1 / (2π·5).

[0062] (7) Figure 2 This is a simulation result diagram of the present invention. The diagram shows a comparison of the position errors of the adaptive error observer method and the traditional method when the system input is an unknown, changing multi-frequency component signal and the disturbance is also an unknown, time-varying multi-frequency component signal. The dotted lines represent the input, disturbance, and signal, illustrating the actual input and disturbance conditions the system handles. The dashed lines represent the error signal of the error observer method, and the solid lines represent the error signal of the adaptive error observer method. It can be seen that in the initial stage (within 15 seconds), the two methods have the same effect. However, when the input and disturbance change (at the 15th second), the system's tracking error suddenly increases, and the tracking accuracy decreases. When the input, disturbance, and signal remain constant for a period of time (after 15 seconds), after the signal identification analysis and Q-filter parameter adjustment process are completed, the tracking error of the adaptive error observer method gradually decreases, while the tracking error of the traditional method remains unchanged. Therefore, it can be seen that the present invention, compared to the traditional error observer method, has a stronger ability to cope with changing inputs and disturbances and has higher tracking accuracy.

Claims

1. A control method for a photoelectric tracking system based on an adaptive error observer, characterized in that, The method includes the following steps: Step (1): Using the CCD as the position sensor of the photoelectric tracking system, the relative angular position between the target's line of sight and the sensor center is calculated, thus forming a position closed loop; Step (2): In the position loop, the object characteristics of the position loop are obtained using a frequency response tester and represented as an object model. The input signal of the frequency response tester is the drive voltage input, and the output signal is the sampled output value of the CCD. Step (3): Based on the object model of the position loop Design a position controller Complete closed-loop control; Step (4): After completing the location loop closure, establish the object model. nominal inverse model The error signal Input to the nominal inverse model , right Output signal and motor drive signal Summation is performed to obtain the input signal in the position loop. and disturbance signals sum and estimate ; Step (5): Identify and analyze using the Fast Fourier Transform method. The main frequency components; Step (6): Construct a Q filter, designing it as a low-pass filter and multiple notch filters connected in series, with variable filtering parameters; Step (7): Based on and estimate the quantity Based on the frequency domain identification analysis results, the Q filter parameters are adjusted to... Perform filtering, and then convert the filtered signal Feedforward Controller Output ; Step (8): Repeat steps (4) to (7) to complete the entire process of the adaptive error observer until the system finishes running.

2. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: In step (2), the object characteristics of the position loop are approximated as a second-order oscillating element, then the object model... The expression is as follows: ， in, , and These are model parameters.

3. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: In step (3), the position controller The zero-pole cancellation method is used for design.

4. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: In step (4), in order to obtain and estimate the quantity Establish an error observer to detect the error signal. Input to the nominal inverse model And the estimate Depend on Output signal and motor drive signal The sum is obtained.

5. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: In step (5), the Fast Fourier Transform method is used to obtain... The power spectrum, and then identification and analysis The main frequency components in.

6. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: In step (6), the Q filter consists of a low-pass filter and multiple notch filters connected in series, and its form is as follows: ， in, , , For filter parameters, For the first The main frequency of a notch filter needs to meet stability requirements. , and .

7. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: In step (7), the stability of the Q filter is constrained by the small gain theorem: 。 8. The photoelectric tracking system control method based on an adaptive error observer according to claim 1, characterized in that: Step (8) includes the following steps: First, construct the error observer structure and obtain the input signal in the loop. and disturbance signals sum and estimate Then obtain the result through the Fast Fourier Transform method. The main frequency components are then used to adaptively adjust the notch filter parameters of the Q filter, and finally the filtered signal is... Feeding forward into the loop enhances the photoelectric tracking system's ability to track unknown changing input signals and its ability to suppress unknown time-varying narrowband disturbances.

Images